Abstract: Recent research on generating 3D assets from single-view inputs by diffusion models has attracted great attention. However, existing diffusion models face challenges such as lack of 3D ...